Investigation of Floods (EDAV Proj.2)

Project Description

In this project, we will analysis the global floods records from 1985 to 2008. Specifically, we will invesitgate the floods patterns over time, floods distribution across continent, causes of floods and relation between floods and air pressure.

Floods patterns over time

First, to have a basic idea about world floods records, we show historical floods in a world map.

Also, a dynamic map is available. To see that interactive map, click the words “dynamic map”.

Floods by continent

Distribution by main cause / death : top floods causes , different from cause of deaths

## Warning: NAs introduced by coercion

Next, we did some analysis of main causes and death. From the distribution of main causes, we found “heavy rain” is the primary main cause of floods, over 2500 registers of floods are caused by heavy rain, which is much higher than any other causes.

However, although “tropical cyclone” and “tidal storm surge” have only around 500 registers, respectively, they were the main causes of death. After ordering the top 10 registers which have the most number of death, 5 of the top 10 registers are caused by these two natural disasters. Compared to the death caused by “heavy rain”, less than 100000 people died in floods caused by “heavy rain”, although it’s the primary main cause of floods.

Displacement

We also wanted to look at how floods affect displacement globally. We are using subset data of countries and respective number of people displaced. In order to get a broad understanding of displacement over time, the graphs have been generated for every 10 years i.e. the number of people displaced has been mapped for the years 1985, 1995, 2005 amd 2015.

From the plots above, we find: + Several regions in African continent have minimal displacement. This may be sttributed to no floods in those regions. This is supported by the finding that the African countries where people got displaced match the countries that had floods. However, it can’t be ruled out that lack of record keeping may be a possible reason.
+ Asia has significant displacement. One of the factors is potentially the highest population density in Asia.

M6 floods caused by heavy rain

From previous analysis, we found that heavy rain is the reason which caused the largest number of floods. Here, we researched on the floods caused by heavy rain that had a level of M6. We count the occurence of M6 level floods at each countries from 1985 to 2008. The countries with the largest 20 occurence are displayed in the following dotchart. From the plot, we can see that in the time period 1985 to 2008 China and USA have much more M6 floods than other countries. We then analysised the M6 floods in China.

In this plot, we plotted all the floods locations with a level at least M6 happened in China from 1985 to 2015 with the death toll. As we can the most deadly floods happened in Hunan province.

How phi data related to floods

Aligned X-axis, ave_phi is increasing， dashed smooth line, M4 and M6 are increasing with some variances, but after 2010, a tendancy of decreasing.

Same tendency as total, but with less magnitude.

Same tendency as total, with higher magnitude.

Based on above plots, we can see that total annual floods have an apparent periodicity. Large floods and extreme floods share a similar trend. What should be concerned is that we are in a dangerous time period now, because in 2015, large floods seem to touch the bottom of its periodicity. Maybe there would be more floods in 2016.

10N-35N has higher average phi, and more floods.
35n-70N has lower average phi, and less floods.

The dashed line is a fiited line from linear regression. We can see apparent positive relationships between average phi and floods whether or not floods are extreme or large.

For extreme floods, it has less variance, while more variance on large floods.